Urban landmark recognition
summary
A mobile app called Alicewho/Lassoo recognizes storefronts and landmarks by matching a 500-byte place signature against the live camera feed, claiming 100 ms recognition today with a path to 10 ms, and reporting 95% accuracy from a single index image and >99% with ten+ images. Indexes come from web crawls plus crowdsourced phone captures, the app caches whole-city indexes on-device, and an SDK/API is planned for licensing to content providers.
AI-written summary · about
chapters (5)
related talks
22mUrban Artificial Intelligence in Amsterdam
10mMapping sidewalks and street furniture with AI and computer vision
22mIsland Mapping: Measuring Urban Footprints with Mapillary & Open Source Tools
18mRecognition and Localization of Parking Signs using Deep Learning
20mWorking with AR by Jayway (Main Stage)
37mSeeing is Believing: Mobile Vision API Deep Dive
20mImplementing Vision AI in Enterprise Android Apps
33mSee More, Be More